Warvito / gradio_medical_ldm

Gradio App - Brain Image Synthesis with Latent Diffusion Models

Instructions

1. Install codecs for ffmpeg

sudo apt install ubuntu-restricted-extras
sudo apt install ffmpeg

2. Create a virtual environment and install the dependencies

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

3. Clone MONAI Generative Models repository and install it

git clone https://github.com/Project-MONAI/GenerativeModels.git
cd GenerativeModels
pip install .
cd ..

4. Download the model weights to the pretrained_models directory The models can be downloaded from here and here.

They should be place in

gradio_medical_ldm/
    L pretrained_models/
        L autoencoder.pth
        L diffusion_model.pth

5. Run the app

gradio app.py

6. Open the app in your browser (tested with Chrome)

Notes

The outputs directory contain the generated .nii.gz images.